Human breast cancer is genetic, in that transformation of breast cells to malignant growth is due to alterations in DNA. For most patients, these alterations are purely somatic, but the same genes may be inherited in altered form in some families. The existence of inherited susceptibility to breast cancer is supported by consistent epidemiological evidence that family history, particularly of early-onset disease, is a significant risk factor for beast cancer worldwide. Therefore, genes responsible for inherited breast cancer can be tracked by linkage analysis of multiple-case families, so long as complexities such as age- and sex-dependent risk, incomplete penetrance, a high background rate of noninherited (sporadic) disease, and possible linkage heterogeneity are included. Current evidence from our group suggests both a gene for early-onset familial breast cancer on chromosome 17q and linkage heterogeneity of the disease. The primary goals of this project are to verify the existence of, close in on, and ultimately identify, a gene responsible for early- onset familial breast cancer. The approach will be to continue gene mapping of breast cancer, both in the families with susceptibility apparently linked to 17q and in additional families with early-onset disease. Markers will include candidate genes from this region of 17q and new polymorphisms from chromosome 17-specific cosmid libraries. Polymorphisms in or near candidate genes will be developed by screening for RFLPs, VNTRs, PCRPs (i.e. RFLPs detectable following PCR amplification of a target sequence), CA-repeats, AluVpA, and polymorphisms in 3' untranslated regions of genes. The candidate genes will then be mapped relative to known 17q markers CEPH families and tested for linkage to breast cancer. In addition, promising candidate sequences will be compared directly between affected members of linked families vs their unaffected relatives vs unrelated controls. Cosmids will be localized to 17q12-q21 by fluorescence in situ hybridization and by screening against partial-deletion somatic cell lines. Polymorphisms mapping to 17q12-q21 will be developed by screening for RFLPs, VNTRs, and CA-repeats. Cosmids containing genes expressed either in breast tumor cell lines or normal tissue (or both) will be identified for especially intensive polymorphism search. Cosmid-based polymorphisms will be mapped in the CEPH families and tested for linkage to breast cancer. Breakpoints on chromosome 17q near genes involved in breast carcinogenesis will be sought by comparing DNA from nonfamilial, early- onset breast cancer patients with congenital developmental disabilities to their unaffected parents, using RFLP analysis and pulsed-field gel electrophoresis to localize de novo chromosomal alterations. There is also some evidence that the estrogen receptor (ER) gene may play a role in inherited later-onset breast cancer. In one family with later- onset breast cancer (not linked to 17q), eight affected women have inherited the same ER haplotype; it is not present in their one unaffected postmenopausal relative. Sequences of ER exons will be compared for the affected women vs their unaffected postmenopausal relative vs unrelated controls. Mapping genes for familial breast cancer is important because alterations at the same loci may be involved in both inherited and sporadic disease. This hypothesis can be tested when the alterations responsible for familial disease are identified. Ultimately, the understanding of genes responsible for familial breast cancer may contribute to very early diagnosis and treatment of breast cancer generally.